1. Field of the Invention
The present invention relates to a device and a program for structured data generation of structuring non-character data such as facsimile data, moving image data and voice data, as well as to a data structure generated by them. Moreover, it relates to a processor and a program for processing a structured document.
2. Description of the Related Art
There are proposed structured document standards such as XML (Extensible Markup Language), SVG (Scalable Vector Graphics), XHTML (Extensible Hypertext Markup Language) and VoiceXML (Voice Extensible Markup Language). Among these standards, XML is most popular. SVG is a standard that is extended XML literally applied to graphic vector data. XHTML is a standard that is redefined HTML by XML, and VoiceXML is a standard that is XML applied to voice data. In these standards, data is limited to character data for structuring by tag and so on, which an information processor can deal with through understanding its semantic content. XML, SVG, XHTML and Voice XML are markup languages for structuring data, and they are different from HTML (Hypertext Markup Language) for expressing data (with the designation of tab, linefeed, font and the like.).
The XML, SVG, XHTML and Voice XML are characteristically capable of defining an attribute name related to a structured data. The attribute is a set of the data name and the data itself, and the data can be interpreted and processed in terms of its name related to its meaning or semantic content. As far as the data processing convenience is concerned, a processable data is limited to character data, and the processable data especially in XLM is limited to non-numeric character data, that is text data.
SVG is a standard for the structured document, as well as SVG (Scalable Vector Graphics) can manipulate vector data. So if the raster format image data is transformed into the vector format image data, SVG can structure the vector format image data so as to easily reduce, expand, eliminate, add and modify the vector format graphic data. Accordingly, VoiceXML (Voice Extensible Markup Language) can manipulate voice data. However, these standards are exclusively aimed at structuring image data and voice data. Actually apart from SVG and VoiceXML, there are ongoing research topics on voice recognition and image recognition which are, for example, finding ways of understanding meanings of voice data and image data, extracting characteristics from voice data and image data, cutting off objects from image data, or extracting movements of objects or persons from moving image data, and so on.
Since, in these standards for structured document, the processable data is limited to text data, non-structured document such as still image data, moving image data or voice data is non-character data, which is not processable.